Isabelle Cudlitz
commited on
Commit
·
f2d191c
1
Parent(s):
73c254c
Upload model
Browse files- README.md +18 -0
- adapter_model.bin +1 -1
README.md
CHANGED
|
@@ -2376,4 +2376,22 @@ The following `bitsandbytes` quantization config was used during training:
|
|
| 2376 |
### Framework versions
|
| 2377 |
|
| 2378 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 2379 |
- PEFT 0.6.2
|
|
|
|
| 2376 |
### Framework versions
|
| 2377 |
|
| 2378 |
|
| 2379 |
+
- PEFT 0.6.2
|
| 2380 |
+
## Training procedure
|
| 2381 |
+
|
| 2382 |
+
|
| 2383 |
+
The following `bitsandbytes` quantization config was used during training:
|
| 2384 |
+
- load_in_8bit: True
|
| 2385 |
+
- load_in_4bit: False
|
| 2386 |
+
- llm_int8_threshold: 6.0
|
| 2387 |
+
- llm_int8_skip_modules: None
|
| 2388 |
+
- llm_int8_enable_fp32_cpu_offload: False
|
| 2389 |
+
- llm_int8_has_fp16_weight: False
|
| 2390 |
+
- bnb_4bit_quant_type: fp4
|
| 2391 |
+
- bnb_4bit_use_double_quant: False
|
| 2392 |
+
- bnb_4bit_compute_dtype: float32
|
| 2393 |
+
|
| 2394 |
+
### Framework versions
|
| 2395 |
+
|
| 2396 |
+
|
| 2397 |
- PEFT 0.6.2
|
adapter_model.bin
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 21020682
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:814aa471d9faeee9288c0b4102ef60a631bea167779ac7f89fe85ea9b4aa70f2
|
| 3 |
size 21020682
|